Performance of Operational AMSR2 Based Sea Ice Concentration Retrieval Algorithms under Extreme Weather Conditions

Satellite products on sea ice concentration (SIC), based on the measurements of the Advanced Microwave Scanning Radiometer 2 (AMSR2), are analyzed in the extreme weather events of Polar Lows (PLs) and Extratropical Cyclones (ECs) in the Arctic, associated with high values of total atmospheric water vapor content (WVC), total cloud liquid water content (CLW) and sea surface wind speed (SWS). The database of PLs and ECs in the Arctic is created for the period of 2015-2018. For this database optical and radar images as well as the sea ice maps from the Norwegian Meteorological Institution and the Arctic and Antarctic Research Institution are collected as ground truth data for SIC in the considered extreme weather events. The two SIC products s - a daily averaged one (Level 3), provided by the University of Bremen and produced with an ARTIST Sea Ice algorithm (ASI), and the SIC product of original temporal resolution (Level 2), produced with a Bootstrap algorithm, provided by the Japan Aerospace Exploration Agency (JAXA), - are searched for the areas of spurious sea ice concentrations (SSIC) in the regions of open water, confirmed by collected ground-truth data. The values of WVC, CLW and SWS are calculated from the AMSR2 data with advanced retrieval methods. Their influence on the SSIC appearance is investigated for both SIC products. It is found that SSIC appears in the product of the University of Bremen mostly due to the atmospheric water whereas in the product of the JAXA - mostly due to high winds. The largest number of SSIC events is observed in the regions of most frequent ECs and PLs. The areas of spuriously detected sea ice in the product of JAXA are much lower in summer, but almost the same in winter as compared with those in the product of the University of Bremen.